Connect with us

Technology

Pangea Unveils Definitive Study on GenAI Vulnerabilities: Insights from 300,000+ Prompt Injection Attempts

Published

12 months ago

on

PALO ALTO, Calif., May 15, 2025 /PRNewswire/ — Pangea, a leading provider of AI security guardrails, today released findings from its global $10,000 Prompt Injection Challenge conducted in March 2025. The month-long initiative attracted more than 800 participants from 85 countries who attempted to bypass AI security guardrails across three virtual rooms with increasing levels of difficulty.

The research comes at a critical time as GenAI adoption has accelerated dramatically across industries, with a majority of enterprises now deploying AI-powered applications that interact directly with customers, employees, or sensitive internal systems. Despite this rapid adoption and integration into business-critical operations, many organizations have yet to implement AI-specific security protocols beyond frontier model defaults.

The challenge generated nearly 330,000 prompt injection attempts using more than 300 million tokens, creating a comprehensive dataset that reveals blindspots in how organizations are currently securing their AI applications.

Key Findings:

Non-Deterministic Security Challenge: Unlike traditional cybersecurity threats, prompt injection attacks exhibit unpredictable success rates due to the non-deterministic nature of LLMs. A prompt injection that fails 99 consecutive times may randomly succeed on the 100th attempt, even with identical content.Data Leakage & Reconnaissance Risk: in addition to the risk of sensitive data leakage and inappropriate responses, an AI application can also be exploited for adversarial reconnaissance purposes to reveal context like what server it’s being run on and open ports it can access.Defense in Depth Necessity: Organizations relying solely on native LLM guardrails are the most vulnerable—approximately 1 in 10 prompt injection attempts succeeded against basic system prompt guardrails. Multi-layered defenses reduced successful attacks by orders of magnitude.Agentic AI Amplifies Risk: As organizations move toward agentic AI with database and tooling access, compromised systems could enable sophisticated lateral movement within networks, dramatically elevating the potential impact of prompt injection attacks.

“This challenge has given us unprecedented visibility into real-world tactics attackers are using against AI applications today,” said Oliver Friedrichs, co-founder and CEO of Pangea. “The scale and sophistication of attacks we observed reveal the vast and rapidly evolving nature of AI security threats. Defending against these threats must be a core consideration for security teams, not a checkbox or afterthought.”

Joey Melo, a professional penetration tester and the only contestant to successfully escape all three virtual rooms, spent two days developing a multilayered attack that ultimately bypassed the single level in room three.

“Prompt injection is especially concerning when attackers can manipulate prompts to extract sensitive or proprietary information from an LLM, especially if the model has access to confidential data via RAG, plugins, or system instructions,” noted Joe Sullivan, former CSO of Cloudflare, Uber, and Facebook. “Worse, in autonomous agents or tools connected to APIs, prompt injection can result in the LLM executing unauthorized actions—such as sending emails, modifying files, or initiating financial transactions.”

In response to these findings, Pangea recommends organizations implement a comprehensive security strategy for AI applications that includes:

Multi-Layered Guardrails: Deploy guardrails to prevent prompt injection, protect the system prompt, prevent confidential information and PII exposure, and detect malicious entities using statistical and LLM-driven analysis techniques.Strategic Attack Surface Reduction: Balance functionality with security by restricting input languages, operations, and response types in security-sensitive contexts.Continuous Security Testing: Implement red team exercises specifically designed to test AI applications against evolving prompt injection techniques.Dynamic Temperature Management: Consider reducing model temperature settings in security-critical applications to minimize randomness that attackers can exploit.Dedicated Security Resources: Allocate one or more resources to track the rapidly evolving prompt injection landscape or partner with commercial security providers specialized in AI defense.

Friedrichs adds, “The industry is not paying enough attention to this risk and is underestimating its impact in many cases, playing a dangerous wait-and-see game. The rate of change and adoption in AI is astounding—moving faster than any technology transformation in the past few decades. With organizations rapidly deploying new AI capabilities and increasing their dependence on these systems for critical operations, the security gap is widening daily. The time to get ahead of these concerns is now.”

The full research report, “Defending Against Prompt Injection: Insights from 300K attacks in 30 days,” is available now, here: https://info.pangea.cloud/prompt-injection-research-report-2025

About Pangea

Pangea’s AI Guardrail Platform empowers security teams to ship secure AI applications quickly and protect workforce AI use with the industry’s most comprehensive set of AI guardrails, easily deployed via gateways or into applications with just a few lines of code. Pangea stops LLM security threats ranging from prompt injection to sensitive data leakage, covering 8 out of 10 OWASP Top Ten Risks for LLM apps, while accelerating engineering velocity and unlocking AI runtime visibility and control for security teams.

For more information, visit pangea.cloud or contact: press@pangea.cloud

Media Contact: Growth Stack Media |  415-574-0738

 

View original content to download multimedia:https://www.prnewswire.com/news-releases/pangea-unveils-definitive-study-on-genai-vulnerabilities-insights-from-300-000-prompt-injection-attempts-302456650.html

SOURCE Pangea Cyber

Related Topics:
Continue Reading
Click to comment

Leave a Reply

Your email address will not be published. Required fields are marked *

Technology

eSign.AI Named Sole Electronic Signature Technology Provider for Hong Kong Government’s CorpID Project, Building the Foundation for Digital Signing Infrastructure in Hong Kong

Published

4 hours ago

on

May 8, 2026

By

HONG KONG, May 8, 2026 /PRNewswire/ — As Hong Kong’s Digital Corporate Identity Platform (CorpID) counts down to its phased launch, eSign.AI has been appointed as the sole electronic signature vendor in the project, responsible for delivering core digital signing capabilities including digital signatures, certificate management, and signature verification services. CorpID is led by Nexify, a seasoned government systems integrator, as the prime contractor. The platform is expected to launch in phases starting late 2026, with multiple CorpID-based e-government services going live in mid-2027.

CorpID: Government-Grade Digital Identity Infrastructure for Hong Kong Enterprises

The Digital Corporate Identity Platform (CorpID) is an enterprise-level digital services platform launched by the Hong Kong SAR Government, developed under the oversight of the Digital Policy Office (DPO). It is designed to serve as the business equivalent of “iAM Smart,” providing a unified digital identity foundation for Hong Kong enterprises. CorpID’s core mission is to build an integrated digital government infrastructure — offering unified identity authentication, digital signing, form pre-filling, and e-licence storage — replacing paper-heavy, cumbersome traditional processes and enabling smart city development through seamless data connectivity.

The platform is open to companies incorporated under the Companies Ordinance (Cap. 622) and businesses registered under the Business Registration Ordinance (Cap. 310), including sole proprietorships and partnerships. The DPO requires all enterprise-related e-government services to support CorpID within 18 months of launch, and will continue expanding ecosystem coverage through sandbox initiatives, cross-industry identity standard interoperability, and fully online registration processes.

eSign.AI: The Digital Signing Engine Behind CorpID

eSign.AI is an AI-native electronic signature and contract automation platform built for enterprises worldwide, offering a complete signing framework from simple electronic signatures to the highest-level compliant digital signatures — meeting diverse regulatory requirements across industries and jurisdictions.

On the identity verification front, eSign.AI has completed integration with iAM Smart, enabling individual identity verification through Hong Kong’s citizen digital identity system, and providing legally valid digital certificate services for both enterprises and individuals.

Looking ahead, the eSign.AI SaaS platform will be deeply integrated with CorpID, providing enterprise and individual identity verification for Hong Kong businesses, and supporting both electronic and digital signing that complies with Hong Kong’s Electronic Transactions Ordinance — connecting the full digital contracting lifecycle for government and enterprise alike.

Getting Ahead of the AI Era: From eSignGlobal to eSign.AI

The electronic signature industry is undergoing a structural shift from “tooling” to “intelligence.” Market data underscores this acceleration: the AI-powered contract analysis tools market has grown from USD 3.32 billion in 2025 to USD 4.3 billion in 2026, at a CAGR of 29.6%. Signing is just one node in the contract lifecycle — document generation, workflow orchestration, compliance tracking, and post-execution management are all being transformed by AI, and the industry window is closing fast.

In April 2026, the company officially rebranded from eSignGlobal to eSign.AI, completing its strategic transformation from an e-signature tool provider to an AI-native contract automation platform. As the company’s spokesperson noted, this rebrand is not cosmetic — it is an acknowledgment of where the product actually is. Customers were already using eSign.AI to automate workflows that go far beyond the signature itself.

eSign Automation Skill was launched alongside the rebrand — an AI-powered signing automation framework for enterprise workflows that enables complete contract signing through natural language interaction, with no manual intervention required. Whether it is single-party approval, multi-party sequential signing, or large-scale parallel execution, an AI Agent can orchestrate the entire workflow in a single call. All signature initiations and status queries return structured JSON outputs, directly parseable by leading large language models and intelligent workflow systems.

eSign Automation is now available in the OpenClaw ecosystem and supports integration via Claude MCP, ChatGPT, and other leading AI platforms.

By combining AI automation capabilities with CorpID’s government-grade digital identity infrastructure, eSign.AI delivers a complete solution for Hong Kong enterprises — from identity verification to intelligent signing to full workflow automation.

About eSign.AI

eSign.AI (formerly eSignGlobal) is an AI-native electronic signature and contract automation platform built for enterprises worldwide. The platform serves over 100 countries and regions, covering core industries including financial services, manufacturing, real estate, human resources, and healthcare — with 1,500+ scenario applications and 3,000+ ecosystem partners. eSign.AI holds ISO 27001, ISO 27701, and ISO 27018 certifications and supports major regulatory frameworks including the U.S. ESIGN Act / UETA, EU eIDAS, HIPAA, GDPR, and 21 CFR Part 11. Infrastructure is anchored by independent data centers in Hong Kong, Singapore, and Frankfurt, Germany.

View original content:https://www.prnewswire.com/apac/news-releases/esignai-named-sole-electronic-signature-technology-provider-for-hong-kong-governments-corpid-project-building-the-foundation-for-digital-signing-infrastructure-in-hong-kong-302766763.html

SOURCE eSignGlobal

Continue Reading

Technology

The 9th AskGamblers Awards Finalists Announced as Voting Starts

Published

4 hours ago

on

May 8, 2026

By

The highly anticipated 9th AskGamblers Awards has officially moved into the voting phase. Following a rigorous selection process, the finalists across 5 premier categories have been revealed: Best Casino, Best New Casino, Best New Slot, Best Sportsbook, Best Provider. Players are invited to cast their votes until 11 June.

BELGRADE, Serbia, May 8, 2026 /PRNewswire/ — The voting stage of the 9th annual AskGamblers Awards has officially begun. The list of finalists is announced, and the first votes are already coming in. 

Players will have a chance to vote for their favourites until 11 June, when the winners will be announced at the gala ceremony in Belgrade. There’s a total of 5 categories where popular votes are taken into consideration:

Best CasinoBest New CasinoBest SportsbookBest New SlotBest Game Provider

There aren’t any big changes to the voting process compared to last year. The votes from the prominent members of AskGamblers Forum will be counted in as well, while some award winners will be announced directly by the AskGamblers teams. 

These include: Best Crypto Casino, Best Partner, and Best Manager categories, while the AskGamblers Superstar Award is expected to be handed to the operator that illustrates the brand values best.

Dijana Radunović, General Manager at AskGamblers, is excited for voting to start: “We’re seeing some familiar contestants, but there are a lot of new names, so it will be exciting to see who comes up on top.”

“We invite players to vote for their favourites! This is a chance for you to speak your mind and support operators and games that shape this industry,” Radunović added.

Before the AskGamblers Awards Ceremony that takes place on 11 June, Charity Night is scheduled for 10 June.

About AskGamblers

AskGamblers.com strives to provide current, objective, and accurate information and guide its users towards a safe gaming experience. The way we deliver our services, from the online casino, sportsbook, slot, and bonus reviews to our trusted Complaint Service, is best described by our motto: ‘Get the truth. Then play.’

For more information about AskGamblers and AskGamblers Awards, please contact dijana.radunovic@g2m.com.

This information was brought to you by Cision http://news.cision.com

https://news.cision.com/askgamblers/r/the-9th-askgamblers-awards-finalists-announced-as-voting-starts,c4346370

View original content:https://www.prnewswire.co.uk/news-releases/the-9th-askgamblers-awards-finalists-announced-as-voting-starts-302766772.html

Continue Reading

Technology

SUNMI Wins 2026 Red Dot Design Awards with Five Products, Leading Global Commercial Industrial Design

Published

4 hours ago

on

May 8, 2026

By

SINGAPORE, May 8, 2026 /PRNewswire/ — The winners of the 2026 German Red Dot Design Award were officially announced. Five of SUNMI Technology’s flagship products won awards: the CPad Business Tablet, CPad PAY, FLEX 3 Interactive Display, the V3 handheld POS Terminal and L3 Industrial PDA. These products stood out with three core design concepts: integration, versatility and human-centricity.

Known as “The Oscars” of global industrial design, the Red Dot Award has strict evaluation criteria covering aesthetics, ergonomics, scenario adaptability and sustainability. SUNMI adheres to original commercial scenario customization, rejecting crudely modified consumer devices. All winning products are originally developed for real commercial scenarios such as cash register, food delivery, industrial inspection and store operations, covering the entire commercial track with high scenario adaptability. Meanwhile, it practices ESG concepts, adopting eco-friendly materials and modular structures to extend equipment service life, reduce consumable consumption, and implement low-carbon and long-term design, which perfectly meets the Red Dot’s sustainability evaluation criteria.

Simplify Complexity: With highly integrated design, SUNMI eliminates the “patchwork feeling” of cluttered devices and tangled cables in traditional commercial scenarios, streamlining store operations and saving space.All-in-One Versatility: Beyond a single tool function, SUNMI’s products achieve flexible transformation through modular and multi-form designs to proactively adapt to changing business needs. The CPad series with modular accessories and FLEX 3’s Lego-style modular design enable multi-scenario application and long-term reuse.Human-Centric Design: Every detail is human-oriented, focusing on real pain points to enhance scenario experience. The L3 Industrial PDA reduces high-frequency work fatigue through scientific weight distribution; the V3 Smart POS Terminal balances large-screen visibility and grip comfort; CPad PAY integrates full-link functions to simplify workflows.

These honors stem from SUNMI’s long-term commitment to a sustainable society, original commercial R&D and ESG. In the future, SUNMI will uphold its core concepts, expand the boundaries of commercial industrial design, and empower global businesses with user-oriented, eco-friendly and high-value products.

Logo – https://mma.prnewswire.com/media/2081156/sunmi_Logo.jpg

View original content:https://www.prnewswire.co.uk/news-releases/sunmi-wins-2026-red-dot-design-awards-with-five-products-leading-global-commercial-industrial-design-302766777.html

Continue Reading

Trending